Cross Random Access Point Sample Groups for Accurate Video Decoding

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video coding standards, such as VVC, lack effective signaling mechanisms for cross Random Access Point (RAP) referencing, leading to potential decoding errors when inter predicted pictures follow DRAP/CRR pictures in decoding order but precede them in output order, and ISOBMFF semantics for DRAP samples are faulty, causing incorrect decoding.

Innovation Solution

Implement a signaling mechanism using SEI messages to indicate whether inter predicted pictures can reference earlier pictures, and enhance DRAP/CRR descriptions in ISOBMFF sample groups to ensure correct decoding order and presentation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If inter predicted pictures are allowed to reference earlier pictures in decoding order, then compression efficiency is improved, but decoding errors occur when pictures follow DRAP/CRR in decoding order but precede them in output order

Engineering Contradiction:

Improvecompression efficiencyVSAvoiddecoding accuracy

Core Design Contradiction:

Loss of energyVSReliability

Solution Approach 1:

The patent introduces a signaling mechanism that预先 (in advance) indicates whether inter predicted pictures can reference earlier pictures. This preliminary signaling prevents decoding errors by establishing reference constraints before the actual decoding process, allowing the system to maintain compression efficiency while ensuring decoding reliability through pre-defined reference picture rules.

Inventive Principle:

Principle #10Preliminary action

2Device complexity

If DRAP sample semantics are implemented without enhanced description, then file format simplicity is maintained, but incorrect decoding occurs due to faulty ISOBMFF semantics

Engineering Contradiction:

Improvefile format complexityVSAvoiddecoding accuracy

Core Design Contradiction:

Device complexityVSReliability

Solution Approach 1:

The patent segments the DRAP sample description into distinct components within the ISOBMFF file format, including separate fields for decoding order, output order, and reference picture information. This segmentation allows the file format to maintain a structured and organized complexity while ensuring that all necessary decoding parameters are explicitly defined, thereby preventing incorrect decoding.

Inventive Principle:

Principle #1Segmentation

3Device complexity

If cross RAP referencing is not signaled, then video coding standard simplicity is maintained, but decoding errors occur in CRR picture scenarios

Engineering Contradiction:

Improvecoding standard complexityVSAvoiddecoding accuracy

Core Design Contradiction:

Device complexityVSReliability

Solution Approach 1:

The patent introduces an intermediary signaling mechanism (SEI message or VPS parameter) that mediates between the simple file format structure and the complex cross RAP referencing requirements. This intermediary layer carries the necessary information about CRR picture relationships without fundamentally complicating the core video coding standard, thereby maintaining decoding accuracy while limiting standard complexity growth.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12407849B2Cross random access point sample group

Publication Date: 2025.09.02 DOUYIN VISION CO LTD

AI summary

A mechanism for processing video data is disclosed. A description of Cross Random Access Point Referencing (CRR) samples is signaled in a visual media data file in International Organization for Standardization (ISO) base media file format (ISOBMFF). A conversion is performed between a visual media data and the visual media data file based on the CRR sample group.